What algorithm did Windows XP use to choose your initial user picture?
Hacker News · 1h ago · 21 points
Programming
The latest Programming headlines, aggregated and deduped from public feeds across the web. See the sources.
Which GPU Do You Need for AI? A Practical VRAM Guide
Learn how to choose the right GPU for LLMs, RAG, image and video generation by comparing VRAM requirements, performance and cost.
HackerNoon · 42m ago
nixamp 0.9: a server that names itself, and a CNN feed that stopped talking
nixamp 0.9: a server that names itself, and a CNN feed that stopped talking A day of nixamp releases, 0.7.36 through 0.9.3. It started with a bug report ("I see CNN but get no audio or video, and the graph is moving") and ended with servers that get their own DNS name and certificate without anyone touching a registrar. CNN had been on the air for twelve hours and showed nobody anything. The graph moved because it was the server's own player mirrored over SSE, not the channel. On the box, the ff
DEV Community · 7m ago
Sources: Moonshot AI is exploring dual Hong Kong and Shanghai listings for greater capital and exposure, amid the weaker performance of AI stocks in Hong Kong (Coco Feng/South China Morning Post)
Coco Feng / South China Morning Post: Sources: Moonshot AI is exploring dual Hong Kong and Shanghai listings for greater capital and exposure, amid the weaker performance of AI stocks in Hong Kong — Moonshot's exploration of dual listings may stem in part from the weaker performance of AI stocks in Hong Kong, the sources said
Techmeme · 7m ago
Why We Abandoned Satellite Imagery for an On-Device Procedural Sky
A working Himawari satellite pipeline produced the wrong view for Amana, so its developer replaced it with an on-device procedural sky.
HackerNoon · 42m ago
Function coloring is real and pretending context propagation isn't infection
Someone recently attempted to make the bold claim that function color isn't a real phenomenon. Unsurprisingly, the internet had a field day with that. An essay went viral on this topic, cross-language, bringing Go, Rust, and Zig into the conflict. The argument: coloring is a non-issue conceptually. Bob Nystrom actually first introduced the concept of "function coloring" in February 2015. His article "What Color is Your Function?" described async functions as red and sync functions as blue (even
DEV Community · 8m ago
DeepSeek debuts DeepSeek-V4.1-Flash, its smallest model built on a new Causal Encoder-Decoder architecture, with 552B backbone parameters and 1M-token context (Reuters)
Reuters: DeepSeek debuts DeepSeek-V4.1-Flash, its smallest model built on a new Causal Encoder-Decoder architecture, with 552B backbone parameters and 1M-token context — Chinese artificial intelligence startup DeepSeek on Thursday launched DeepSeek-V4.1-Flash, which the company said is the smallest model …
Techmeme · 32m ago
I Built a Voice-Based Daily Reflection Companion under 15 Minutes Using Agora Agents SDK
I expected to spend a weekend on working on my vision. Instead, I had a voice agent asking me "How was your day?" in under 15 minutes. What surprised me most wasn't the speed - it was that when I interrupted the voice AI agent I named "Compass" mid-sentence to change my answer, she just stopped and listened. No stuttering, no doubled audio, no ghost speech finishing in the background. It just worked, out of the box, without a single line of interruption-handling code on my end. If you've tried t
DEV Community · 10m ago
Startup Founder Interview: SparkVox Turns Founder Recordings Into LinkedIn Content They Sound Like
SparkVox, featured on HackerNoon, turns founder recordings from audio into authentic LinkedIn posts using AI trained on each founder's voice & editorial style.
HackerNoon · 1h ago
OpenAI have no mathematicians capable of understanding what they put out
Hacker News · 5h ago · 27 points
Letter: the Senate disaster management subcommittee, led by Sen. Josh Hawley, is probing OpenAI's handling of the Hugging Face breach, calling it "reckless" (Axios)
Axios: Letter: the Senate disaster management subcommittee, led by Sen. Josh Hawley, is probing OpenAI's handling of the Hugging Face breach, calling it “reckless” — A Republican-led Senate subcommittee that oversees disaster management is investigating OpenAI's handling of the Hugging Face breach in July, Axios has learned.
Techmeme · 42m ago
Claude Fable 5.1 + GPT-6 Astra: Two Coding Agents, One Workflow
A coding agent can write a patch, explain why it is correct, and then review its own work. What I wanted was a second agent that could investigate the same problem before seeing the first agent's explanation—and then challenge it with evidence. So I built OMP Tandem: an open-source bridge that gives your coding agent an independent AI peer through Oh My Pi. One example pairing is Claude Fable 5.1 in Claude Code, with GPT-6 Astra as the OMP peer. You keep the coding environment you already use, c
DEV Community · 12m ago
Remotely access Home Assistant via Tailscale for free
Learn how to set up Tailscale with Home Assistant for secure remote access using Serve, HTTPS, exit nodes, and subnet routing without opening ports.
HackerNoon · 1h ago
Uptime Kuma: Why HTTP 200 Can Hide a Broken App
Uptime Kuma can show UP while your endpoint reports a failed dependency. If the endpoint still answers HTTP 200, add a check for the specific response field that matters. A broad keyword such as ok can miss the failure too. The example below uses four small test responses and records what 12 real Uptime Kuma monitors reported. Use it before trusting the green badge on a self-hosted app or a service you just shipped. Source checks and local experiment: September 9, 2026. Tested with Uptime Kuma 2
DEV Community · 13m ago
How Spynn Is Getting Founders Into Forbes Without A Publicist
Spynn offers founders guaranteed placement in publications such as Forbes without traditional PR retainers, shifting the risk from clients to the agency.
HackerNoon · 2h ago
A look at why the oft-discussed predictions that AI will deliver double-digit GDP growth in advanced economies are extremely unlikely over the next 10-15 years (Ghosts of Electricity)
Ghosts of Electricity: A look at why the oft-discussed predictions that AI will deliver double-digit GDP growth in advanced economies are extremely unlikely over the next 10-15 years — The following post represents the personal views of the author(s) and does not reflect or represent the positions of their employers.
Techmeme · 3h ago
DelegatingHandler Magic: Build a Pipeline for Your HTTP Calls
What's up, .NET devs! 👋 Today I want to show you one of the most underrated features in HttpClient — the DelegatingHandler pipeline. It's like middleware for your HTTP calls! Think of DelegatingHandler as a chain of interceptors. Every HTTP request passes through each handler, and every response comes back through the same chain (in reverse). 🔗 Fun Fact: The handler pipeline is modeled after the "Chain of Responsibility" design pattern from the Gang of Four book! Request: Client → Handler1 → H
DEV Community · 13m ago
What Happens to Your Data After You Hit Allow
Tapping allow is just the start. Learn what really happens to your data afterward, from storage and reuse to third party sharing and deletion rules.
HackerNoon · 3h ago
The AI boom is fueling a resurgence in VC bets in "moonshot" sectors such as BCI; Dealroom says non-AI deeptech funding has topped $150B since the start of 2024 (Tim Bradshaw/Financial Times)
Tim Bradshaw / Financial Times: The AI boom is fueling a resurgence in VC bets in “moonshot” sectors such as BCI; Dealroom says non-AI deeptech funding has topped $150B since the start of 2024 — Tech investors are rediscovering the kind of long-shot sci-fi bets that helped build Silicon Valley.
Techmeme · 3h ago
App Store IPA Upload Automation: .p8 Key Auth and CI/CD Integration in Practice
On release day, the most annoying step is the manual upload: the build machine produces the package, but a human still has to open the upload tool, pick the file, enter credentials, and watch the progress bar—then do it all over again if the network hiccups. Worse, these operations live on individual laptops, so handing off to someone else means walking them through the process again. To wire "upload the IPA" into CI/CD, the first thing to solve is actually authentication—letting a machine in th
DEV Community · 16m ago
Introducing "Token Proof of Work"
A proposal for stopping AI agent spam without CAPTCHAs — by making every submission cost real tokens and real time, Bitcoin-style.
HackerNoon · 3h ago
Sources: Grab is in talks to buy a majority stake in Advance Intelligence Group's Singapore-based BNPL platform Atome Financial at a $2B+ valuation (Bloomberg)
Bloomberg: Sources: Grab is in talks to buy a majority stake in Advance Intelligence Group's Singapore-based BNPL platform Atome Financial at a $2B+ valuation — Grab Holdings Ltd., the biggest ride-hailing and food-delivery firm in Southeast Asia, is in talks to buy a majority stake in Singapore-based buy …
Techmeme · 4h ago
What Zero Dependencies Actually Meant for RepoX-Ray
## What Zero Dependencies Actually Meant for RepoX-Ray When we started building RepoX-Ray for Zero Dependency 2026, I thought zero dependency mainly meant one thing: Don't install any packages. I quickly realized it meant much more than that. RepoX-Ray is a CLI tool that scans repositories for security, dependency, Git, configuration, and code-quality issues. Normally, we would probably reach for packages like GitPython, Click, Pydantic, or pytest. Instead, we had to ask: “What can Python's stan
DEV Community · 16m ago
Why TDE SQL Server Restores Fail on Amazon RDS
Learn why TDE-encrypted SQL Server backups fail on Amazon RDS and how to migrate certificates correctly using S3, KMS, and RDS procedures.
HackerNoon · 3h ago
How Amazon's Zoox, which has ~100 AVs in its fleet, is utilizing social media influencers, community events, and ad campaigns to compete with Waymo in SF (Emmy Martin/New York Times)
Emmy Martin / New York Times: How Amazon's Zoox, which has ~100 AVs in its fleet, is utilizing social media influencers, community events, and ad campaigns to compete with Waymo in SF — The Amazon-owned driverless car company is a distant second to Waymo. Its pitch to riders? Wine pop-ups, festival sponsorships and a car built to be filmed.
Techmeme · 4h ago
Chronos-2, CLIP, and MobileNetV3: three production AI models for non-text workloads
When I sort the model catalog at aiappdex.com by total downloads, the 15th most downloaded model processes time-series data. The 14th handles zero-shot image classification. The 16th runs on hardware with under 3M parameters. None of them generate text. Three models from that tier deserve attention — not because they're new, but because the download numbers say someone is shipping them in production: Model Downloads Task Chronos-2 25,699,844 time-series-forecasting CLIP-vit-base-patch32 19,936,7
DEV Community · 16m ago
Chatbots Remember Facts. I’m Building One That Remembers Becoming
Evo explores developmental AI memory, tracking how humans and AI companions change together over time, not just what they remember.
HackerNoon · 3h ago
Turn a Repeated Mistake Into a Check
A reminder lives for one session, a rule lives while it is read, a check lives in the tooling. 👋 I'm Anton - a software engineer working mostly in PHP/Symfony and Go, currently carving a live github.com/brilliant-almazov. The thought I want to share is small and slightly uncomfortable: the only place a rule reliably As before: these are my habits on one codebase, not advice for yours. The mistake is boring. Someone writes, by hand, a copy of something the tree already has as a Not "similar func
DEV Community · 16m ago
Chinese tech giants are hiring skilled professionals as specialized AI trainers to build high-quality datasets, mirroring efforts by US platforms like Mercor (Viola Zhou/Rest of World)
Viola Zhou / Rest of World: Chinese tech giants are hiring skilled professionals as specialized AI trainers to build high-quality datasets, mirroring efforts by US platforms like Mercor — Squeezed by a stagnant economy and state directives, China's underemployed lawyers, architects, and engineers are taking cheap gig work …
Techmeme · 4h ago
I Built a Dream Journal for My Oura Ring — Here's What an n=1 Sleep Experiment Looks Like
I connected Oura sleep data with a dream journal to explore whether dreams, HRV, REM sleep, and recovery reveal useful personal patterns.
HackerNoon · 3h ago
How systematic review catches the GitHub Actions assumptions authors can't see
The conclusion I reached this week: when I changed my YouTube video pipeline from a daily cadence to three times per week, a Codex review found four defects. A second review, run the next day after I fixed those four, found four more. All eight fell into the same category. I had built assumptions about external system behavior directly into the code, and I couldn't see them because I built the system knowing how it was supposed to work. A reviewer who doesn't share that knowledge can ask questio
DEV Community · 16m ago
Q&A with AI researcher Jacob Coxon, who quit Anthropic, on the need for industry-wide, international coordination to limit recursive self-improvement, and more (Maxwell Zeff/Wired)
Maxwell Zeff / Wired: Q&A with AI researcher Jacob Coxon, who quit Anthropic, on the need for industry-wide, international coordination to limit recursive self-improvement, and more — Jacob Coxon talks to WIRED about the “mini Manhattan project” inside Anthropic, the problem with alignment …
Techmeme · 4h ago
The Absurdity of High Commissions: Why Marketplaces Need to Move Beyond the Transaction Fee
High marketplace commissions are creating real conflict. Alexey Grushko explains why Marketplace 3.0 may move beyond transaction fees.
HackerNoon · 3h ago
macOS has no public API for screen brightness
Try to write a Mac app that changes the screen brightness. Not reads it — changes it. You will find nothing in AppKit, nothing in Quartz, nothing in ScreenCaptureKit. NSScreen will tell you a display's colour space, its frame, its refresh rate, and its maximum EDR headroom. It will not let you dim it. This is not an oversight that has gone unnoticed for a year. It is the state of the platform, and working around it teaches you something about what "private API" actually costs. On Intel Macs the
DEV Community · 18m ago
Sources: Chinese AI chipmakers Huawei, Cambricon, MetaX, and Iluvatar CoreX have raised prices for current and next-gen chips by 20%-50% due to rising HBM costs (Reuters)
Reuters: Sources: Chinese AI chipmakers Huawei, Cambricon, MetaX, and Iluvatar CoreX have raised prices for current and next-gen chips by 20%-50% due to rising HBM costs — Chinese AI chipmakers including Huawei Technologies [RIC:RIC:HWT.UL] and Cambricon (688256.SS) have sharply raised prices …
Techmeme · 4h ago
When Random Bugs Point to a Bigger Product Problem
Repeated bugs, flaky tests, and support issues may point to the same weak area. Learn why teams should map patterns instead of only closing tickets.
HackerNoon · 3h ago
My reproducible Erlang playground with Nix
I started learning Erlang and wanted somewhere to keep the small programs and experiments I write along the way. Since I also prefer to manage my project's system dependencies with Nix, I created a small Nix development shell to pin them. For now, it contains little more than Erlang itself. The repository for the playground is available on GitHub. There are two main parts: flake.nix, which defines the Nix development environment lyse/, which contains a few programs I've written while working thr
DEV Community · 25m ago
Sources including French public officials detail how Mistral's rise over the past three years owes much to the privileged relationship it built with Macron (Océane Herrero/Politico)
Océane Herrero / Politico: Sources including French public officials detail how Mistral's rise over the past three years owes much to the privileged relationship it built with Macron — Mistral has forged unrivaled ties with the French presidency. Now, Emmanuel Macron is departing. — in Paris
Techmeme · 5h ago
Your AI Video Model Just Got Retired. Here's How to Keep Your Pipeline From Breaking Next Time.
Runway retired two AI video models overnight in 2026. The habits that keep an AI filmmaking pipeline from breaking when a provider updates its models.
HackerNoon · 3h ago
Machine Learning for Customer Churn Prediction: What Actually Works
Some of your customers are leaving because you called them A retention call reminds someone their contract is ending. It prompts a price comparison they had not thought to run. It reopens a complaint they had let go. For a slice of your base, being contacted is what triggers the exit. That group sits on the same list as everyone you intend to save, and a model ranked by churn probability cannot separate them out. This is the flaw at the centre of most churn programmes, and it is not a modelling
DEV Community · 25m ago
OpenAI Foundation board member Paul Christiano says the AI industry is currently not on track to reduce the acute loss-of-control risk to an "acceptable" level (Paul Christiano/@paulfchristiano)
Paul Christiano / @paulfchristiano: OpenAI Foundation board member Paul Christiano says the AI industry is currently not on track to reduce the acute loss-of-control risk to an “acceptable” level — I am excited to be joining the OpenAI nonprofit board, serving on the Safety and Security Committee to support safety oversight.
Techmeme · 5h ago
Release Engineering for AI Features: Rollouts, Rollbacks, and Runtime Risk
AI features change without code changes. Learn how release teams can manage staged rollouts, rollback paths, model drift, and runtime risk safely.
HackerNoon · 3h ago
Seat-based AI tooling has an idle capacity problem nobody budgets for
If your company buys per-seat AI coding subscriptions, you are almost certainly paying for a lot of capacity that expires unused every week, while some of your engineers hit their limit on Tuesday. Both things are true at once, and that combination is what makes it a distribution problem rather than a spending problem. Usage is bursty and it correlates with role in a way that seat allocation does not. Someone mid-migration will saturate a subscription for two weeks and then barely touch it for a
DEV Community · 27m ago
Sources: Alibaba is set to lead a $300M round in AI model testing startup UniPat AI at a $2.5B valuation; UniPat founder Li Kuan worked at Alibaba's Tongyi Lab (Bloomberg)
Bloomberg: Sources: Alibaba is set to lead a $300M round in AI model testing startup UniPat AI at a $2.5B valuation; UniPat founder Li Kuan worked at Alibaba's Tongyi Lab — Alibaba Group Holding Ltd. is slated to lead a $300 million investment in AI training and benchmarking startup UniPat AI …
Techmeme · 6h ago
DevOps Metrics That Actually Matter: DORA, MTTR, and Beyond
Discover the key DevOps metrics that truly drive success, including DORA, MTTR, and more.
HackerNoon · 3h ago
Regímenes de mercado con Hidden Markov Models: un score de riesgo que bate al buy-and-hold
Los mercados no se comportan igual todo el tiempo. Hay periodos de calma alcista, periodos neutrales y periodos de pánico. El problema es que ese "régimen" es un estado oculto: no viene etiquetado en los datos, solo lo intuimos por cómo se comportan los retornos y la volatilidad. Es el escenario perfecto para un Hidden Markov Model. Quería un score de riesgo en tiempo real que no se limitara a mirar la volatilidad pasada, sino que identificara en qué régimen está el mercado ahora y ajustara la e
DEV Community · 27m ago
Amazon names Mandiant founder Kevin Mandia to its board of directors, noting his extensive experience "combating cyber threats" in public and private sectors (Annie Palmer/CNBC)
Annie Palmer / CNBC: Amazon names Mandiant founder Kevin Mandia to its board of directors, noting his extensive experience “combating cyber threats” in public and private sectors — Amazon said Kevin Mandia, the founder and former CEO of cybersecurity company Mandiant, is joining its board of directors.
Techmeme · 6h ago
OpenSpec Wasn’t Enough: How AIDLC Fixed Our AI Coding Workflow
AI made coding faster but debugging worse. Learn how AIDLC helped one team improve code quality with dynamic workflows and human approval gates.
HackerNoon · 3h ago
Data Quality Checks That Catch Real Web Scraping Failures
A scraper can run successfully and still give you bad data. The HTTP request returns 200, your parser finds elements, the job writes rows, and nobody notices that price now contains the text from a promo banner because the site changed a CSS class. That is the annoying part of scraping: failure often looks like success until someone uses the dataset. The first mistake is storing whatever the page gives you and cleaning it later. That works for one-off scripts. It breaks down when you run scraper
DEV Community · 27m ago
AI Made Coding Faster—So Why Did Software Delivery Get Slower?
AI sped up coding, not delivery. Here is where the bottleneck actually moved, and what engineering leaders should measure instead of coding speed.
HackerNoon · 3h ago
6 Best AI Pentesting Tools in 2026
AI pentesting tools are changing how organizations validate security weaknesses. Instead of relying only on periodic manual assessments or vulnerability scanners, security teams can now use autonomous or agentic systems to test applications, APIs, infrastructure, identity, and cloud environments more frequently. The strongest platforms go beyond detection. They attempt exploitation, verify findings, reproduce attack paths, and help teams confirm remediation worked. But these products are not int
DEV Community · 30m ago
Seven Sign-Ups, Zero Catches: What My Fishing App Launch Taught Me
We launched in 177 countries and nothing happened. Seven sign-ups, all gone within a day. The gate log showed they were not idle - they were refused.
HackerNoon · 4h ago
What Actually Changed for Bots
GitHub: https://github.com/abrownfox0/abrownfox001-twap60-prediction-trigger-system YouTube walkthrough: https://www.youtube.com/watch?v=XzhugRL6BV4 Live profile: https://polymarket.com/@abrownfox001 This is part 1 of a 5-part series on the latest Polymarket platform changes and the tech work they force on short-horizon bots. Part 1 — The change map (this post) Part 2 — TWAP60 settlement: what a 5m engine must pin Part 3 — Taker delay, rate limits, and execution reality Part 4 — Combos, APIs, an
DEV Community · 32m ago
NeoHorse-1-4B: A 4B Model Built for AI Agents and Tool Use
Explore NeoHorse-1-4B, a Qwen3.5-based 4B model built for AI agents, tool use, coding, instruction following, and long-context tasks.
HackerNoon · 4h ago
Our MCP server was fine. Cloudflare was returning HTML.
This is what our MCP endpoint returned to a client speaking JSON-RPC: HTTP/2 403 content-type: text/html; charset=UTF-8 server: cloudflare Attention Required! | Cloudflare Please enable cookies. Sorry, you have been blocked A program was being asked to enable cookies. We did not see that for three days, because that is not what the client reported. The client reported a parse failure, and a parse failure points at your own serialization, not at a machine four thousand kilometres away. The connec
DEV Community · 37m ago
Silent HMAC Key Contamination: Uncovering a Logic Flaw in Burp's JWT Editor Extension
JWT Editor was shortlisted for “Best Auth & Access Control” in PortSwigger’s 2026 Burp Suite Extension Awards. This is the story of finding a silent bug inside.
HackerNoon · 4h ago
An AI-generated login endpoint "works" - I still found SQL concatenation before launch
Functional tests passing is not the same as code being safe to ship. Here is a reproducible case: an AI-generated login endpoint that behaves correctly, the scan finding I checked before launch, and the fix that made the pattern disappear. This is a local demo equivalent of a login endpoint, not live production code and not a client project: username = request.form.get("username", "") password = request.form.get("password", "") sql = f"SELECT id FROM users WHERE username = '{username}' AND passw
DEV Community · 40m ago
A release manifest for generated audio and ad images
Generated media needs a release record just as much as application code does. When a team cannot tell which script produced an audio file, or which approved offer belongs to an image, the problem usually appears during review rather than generation. Disclosure: This article was created with AI assistance. It describes a proposed engineering workflow, not measured results from a production deployment. I am an independent ElevenLabs and AdCreative.ai affiliate; the optional links at the end may ea
DEV Community · 41m ago
How I Built a Local Silence Remover That Processes a 5GB Audio File in Under 6 Minutes
VoxCut is a dedicated silence remover for audio and video: it automatically detects and cuts silent pauses — dead air, long breaths, gaps between takes — from podcasts, interviews and recordings. Unlike full editors that bundle silence removal as one feature among many, it does this one job, entirely on your device. I want to walk through the actual engineering problem behind it, because "remove silence from audio" sounds trivial until someone hands you a 72-hour, 5GB recording and expects it ba
DEV Community · 43m ago
Build a Keyword-Based Resume Analyzer with FastAPI and React
Job seekers often compare their resumes with job descriptions to understand which technical skills overlap and which ones are missing. Doing this manually becomes repetitive when applying for several roles. In this tutorial, you will build a full-stack application that performs this comparison using transparent, rule-based keyword matching. The application lets a user sign up, log in, upload a PDF resume, and paste a job description. It extracts recognized technical skills from both texts and re
DEV Community · 43m ago
5 common TypeScript interview questions
These five questions show up in almost every TypeScript interview. They're not tricky syntax puzzles. They test whether you understand the type system well enough to use it on purpose instead of just making the compiler happy. any vs unknown any turns type checking off. Once a value is any, TypeScript stops protecting you and mistakes slip through until runtime. unknown is the safe counterpart: it can hold anything, but you must narrow it before using it. const a: any = "hello"; a.toUpperCase();
DEV Community · 47m ago
codsh 0.22.1 — /ship grill cards wrap long questions (DeepSeek terminal agent)
DeepSeek-native terminal coding agent; /ship one sentence → verified code. 0.22.1: grill card wraps long questions fully (no 2-line ellipsis clip); sync dsh packages to 0.1.5-rc.1 Install: npm i -g @deepseek-ai/dsh codsh-cli && codsh Links: GitHub · Docs · npm codsh-cli Not a Claude Code env wrapper. Feedback and a star are always appreciated.
DEV Community · 47m ago
A futuristic Bloomberg-inspired smart-money intelligence terminal for Solana wallets
🪐 Smart Wallet Terminal Track influencers, smart-money wallets, top profit leaders, and whales from one animated terminal dashboard. Features • ** Smart Wallet Terminal** is a Python terminal application that transforms 's smart-money wallet feeds into a live, keyboard-controlled market intelligence dashboard. The interface is designed around the visual language of professional financial terminals: High-density wallet tables Bright market-status colors Animated loading and printing effects Live
DEV Community · 48m ago
AI Dev Weekly #25: GPT-6 Astra Arrives, Kotlin Agents Reach 1.0, Copilot Adds Enforced Permissions
AI Dev Weekly is a Thursday series where I cover the week's most important AI developer news, with my take as someone who actually uses these tools daily. Four different layers of the agent stack changed this week. OpenAI introduced GPT-6 Astra for the hardest tool-rich work. Google made its Kotlin agent framework production-ready. GitHub gave enterprise administrators permissions that local settings cannot weaken. And NVIDIA released a local inference router that spreads independent agent calls
DEV Community · 49m ago
AI agents in Rails, for a business that actually has customers
Originally published on gkosmo.eu. An agent is a background job that gets to call a few methods I have been putting this post off for months. Partly because "AI agents" is a phrase that makes me tired, and partly because most of what I read about it is either a demo that writes haikus or an architecture diagram with eleven boxes. Here is what I actually run. It's small. You can paste it into a Rails app this afternoon. The gems are rcrewai and rcrewai-rails. I maintain both and I use them on nak
DEV Community · 52m ago
Designing a Testable Image-to-Video Prompt Workflow
Designing a Testable Image-to-Video Prompt Workflow For a browser-based starting point, Photogenerator.ai brings image and video creation into one workspace. AI video interfaces often look simple: upload an asset, enter a prompt, click generate. The hard part is not the button. It is defining a contract for what the input controls, what the prompt controls, and what the reviewer is expected to accept. This post is an interface-level design memo. It does not infer a vendor's source code, backend,
DEV Community · 53m ago
Mathematicians, OpenAI, and a $1m Problem
Two proofs went public this week. The dispute is about affiliation, not mathematics. The Navier-Stokes equations mathematically describe how fluids move, and there's been a $1m bounty on proving their solutions since 2000. Twenty-six years later, those tough math problems became the subject of a complicated discussion on ownership in the age of AI. On September 8, Tristan Buckmaster, Levent Alpöge, and Matei Coiculescu published proofs showing that several closely related equations, including 3D
DEV Community · 54m ago
Building Perisai: the app that does the maths your lender hopes you won't
A story about online loans, one very confused calculator, and five apps pretending to be someone they're not. Imagine you borrow Rp1.000.000 from an app on your phone. That's the number on the contract. But when the money lands in your account, it's only Rp800.000. The app kept Rp200.000 as an "admin fee." Fine, whatever. Thirty days later, the app says you owe Rp1.300.000. Most people look at that and think: steep, but I'll manage. Here's what it actually is. You received Rp800.000 and paid bac
DEV Community · 54m ago
Set up Astro with Svelte, Tailwind CSS, Vercel, Prettier, and TypeScript
I use Astro with Svelte for interactive components, Tailwind CSS for styling, and Prettier to keep formatting consistent. This guide walks through the project setup, strict TypeScript checks, and a static deployment to Vercel. Before you begin, make sure you have the following on your machine: Node.js 22.12.0 or higher (odd-numbered releases are not supported) Visual Studio Code with these extensions: astro-vscode, prettier-vscode, svelte-vscode, and tailwindcss-intellisense That's it. Astro nee
DEV Community · 55m ago
Your coding agent finished. Now find the PR, the preview, and the right session.
The agent says it fixed the checkout bug. You still need to find the pull request, read the diff, check the build, and open the right preview. Tomorrow, you need to remember which conversation contains the fix. That last part is easy to underestimate. A useful agent session is a record of decisions, failed attempts, commands, and follow-up questions. Losing it means reconstructing work you've already done. I built Cogpit to keep that work together. It's a free, open-source GUI for Claude Code, O
DEV Community · 1h ago
Stop spawning an MCP server per agent session (and what it won't fix)
Ten parallel Claude sessions. Ten copies of the same MCP server. Ten processes, ten sockets to the same upstream, ten holders of the same lock — because that is what stdio means. I moved every server to one shared daemon per machine bound to 127.0.0.1, and the fleet stopped fighting itself. TL;DR: an MCP server registered as stdio is spawned per client session. Register it as an HTTP/SSE URL instead and every session shares one process. It saves memory, sockets and locks. It does not save tokens
DEV Community · 1h ago
Turning a list of company websites into a contact list with a plain HTTP crawl
Anyone who has done outbound sales or recruiting knows this task. You have a list of company websites and you need the contact info off each one. Open the Contact page, copy the email, check LinkedIn, move to the next site. A 50 domain list eats most of an afternoon done by hand. Give it a start URL and it crawls within that same domain, up to a depth you choose, using Crawlee's CheerioCrawler. On each page it checks for: Email addresses in the raw HTML Phone numbers Social profile links (Linked
DEV Community · 1h ago
7 Free Jira Alternatives for Software Teams in the Agent Era
I’ve used enough project management tools to know that the hardest part usually isn’t creating a board. It’s keeping the board useful once the project gets busy. Tasks multiply, priorities change, developers open issues, product requirements evolve, and eventually the project management tool becomes another thing the team has to maintain. Jira has been around this problem for a long time. It gives software teams a structured way to manage issues, sprints, backlogs, workflows, and releases. But s
DEV Community · 1h ago
The AI Race Is a Governance Problem Before It Is a Superintelligence Problem
This week, a researcher who said he spent three years working at OpenAI and Anthropic resigned publicly and warned that leading labs are moving too quickly toward systems that may become difficult to control. His warning is his own assessment, not a prediction anyone can prove today. But it raises a question that engineering teams should not wait to answer: What evidence do we have of what our AI systems are actually doing? The long-term debate is about increasingly capable and autonomous system
DEV Community · 1h ago
App Store Submissions Now Open for iOS 27, macOS 27, and Other OS Releases
Apple has announced that developers can now submit their apps and games to the App Store, leveraging the new features and innovations in iOS 27, iPadOS 27, macOS 27, tvOS 27, visionOS 27, and watchOS 27. These updates include capabilities like Apple Intelligence and the Foundation Models framework. Developers are encouraged to download the Xcode 27 Release Candidate, build their applications with the latest SDKs, test them using TestFlight, and then submit for review. A key change for macOS deve
DEV Community · 1h ago
Stage 1 of enterprise AI adoption: identity and the AI gateway
This is part 1 of a series that started with the full reference diagram in part 0. Stage 1 covers the two columns marked 1 on that diagram. The first is IDENTITY & ACCESS, with SSO / IdP, Roles & scopes, and Secrets manager. The second is AI GATEWAY, with Routing, Rate limit & budget, Logging, and PII / DLP filter. Cloud LLM APIs sit on the enterprise boundary line, and every arrow to them passes through the gateway first. This stage comes first because each later stage assumes two questions are
DEV Community · 1h ago
Your tracker has no private draft state, and it shows
Every issue tracker I have used has some version of Backlog, In Progress, Done. Not one of them ships a place to prepare a task in private before other people can see it. That missing state causes a specific, recognisable mess, and once you notice it you see it everywhere. Somebody has an idea. The only way to record it is to create a ticket. Creating a ticket puts it on the board. Being on the board means it is now a thing the team can see, comment on, estimate, and start. So a half-formed thou
DEV Community · 1h ago
The delegation test is whether you can write the check first
Every discussion I see about what to hand off to a coding agent sorts tasks by difficulty. Easy things to the agent, hard things to the human. It sounds obvious and it has been wrong in practice for me almost every time. The axis that actually predicts a good handoff is not difficulty. It is this: can you write down, before any work starts, what would make you accept the result? If you can write the check, hand it off. If you cannot, the task is not ready to hand to anybody. I have handed off ge
DEV Community · 1h ago
How Much Should You Prepare for an Interview? A Time Budget
The usual answer is "five to ten hours." It is not wrong, and it is not useful, because it never says what those hours are spent on — so people fill them with the easiest activity available, which is reading the company's About page a fourth time. The number is the wrong thing to ask about. Most of what makes you good in an interview is prepared once and reused for every interview after it. Once you separate the one-time work from the per-interview work, the question stops being "how many hours"
DEV Community · 1h ago
SENTINEL-RL for SOCs: Architectural Gains and Cost Realities from Decoupling Semantic and Topological Reasoning
SENTINEL-RL for SOCs: Architectural Gains and Cost Realities from Decoupling Semantic and Topological Reasoning Security operations centers (SOCs) hit scaling limits when authentication graph analysis jams both semantics and topology through a single bottleneck. Most toolchains intertwine context processing, action selection, and graph traversal tightly enough that tuning for scale or specialization is impossible. The result: wasted human cycles, runaway costs, and routine breakdowns in multi-th
DEV Community · 1h ago
Day 15: Visualizing Embeddings with t-SNE and PCA
Previously, on Day 14: Explained how word embeddings represent words as vectors, the limitations of classic approaches like one-hot encoding, and how FastText uses subword units to create robust embeddings that handle rare, new, or misspelled words. Word embeddings are dense vectors that represent words as points in space. Imagine each word having a "position" in a space with maybe 100 or 300 directions—far beyond our normal three. Each number in the vector says how much the word lines up with o
DEV Community · 1h ago
Magento 2 RabbitMQ Performance: Tuning Consumers and the Broker for High-Volume Stores
RabbitMQ is the invisible engine behind Magento 2's asynchronous work: bulk REST operations, async endpoints, order emails, product alerts, inventory reservation cleanup and B2B quote and shared-catalog updates all travel through message queues. The storefront can be perfectly fast while the queues silently back up — orders confirmed, but emails arriving an hour late, bulk operations stuck on "processing", and inventory_reservation rows growing because the cleanup consumer never caught up. This
DEV Community · 1h ago
Bounties for AI work have the same failure mode as game economies
I spent a few years designing game economies before I spent any time on AI tooling, and the transition has been strange, because I keep watching people rediscover failure modes that mobile games documented years ago. The current one: bounties for AI-assisted work. Pay people for tasks completed by their agent. Put a leaderboard on it. Watch throughput go up. Throughput does go up. That is the problem. Every reward loop has a currency, a source, and a sink. The loop stays healthy while the effort
DEV Community · 1h ago
Vision Over Syntax: I Design the Entire Product in My Head Before the First Commit
[ EXECUTIVE TEARDOWN // TL;DR ] Every shipped product started as a complete mental model — data flow, trust boundaries, failure modes, cost curve — before any code. The strongest architecture decisions are refusals: what the system will never do deletes entire classes of future cost and risk. AI made implementation nearly free, which makes knowing what to implement the highest-leverage skill in engineering. The thinking phase runs on four artifacts: a data contract, named trust boundaries, a cos
DEV Community · 1h ago
Halving WebSocket payloads with MessagePack in Node and React
[ EXECUTIVE TEARDOWN // TL;DR ] Batching messages beats changing the encoding; it costs nothing and usually wins more. Set binaryType to arraybuffer on the browser socket, or every message pays for an async Blob read. MessagePack lands 40 to 55 percent smaller than JSON for telemetry-shaped objects but still encodes the keys. Measure with permessage-deflate on and off; compressed JSON sometimes beats uncompressed binary. A telemetry socket sending 40 small JSON messages a second works fine on a
DEV Community · 1h ago
ERC-8403 Standardizes Account Authority, Vitalik on EIP-8141, Base Vibenet Tests 200ms Blocks, Robinhood Chain’s Blobs Stall
Welcome to our weekly digest, where we unpack the latest in account and chain abstraction and the broader infrastructure shaping Ethereum. This week: a new standard defines the lifecycle for how native AA accounts add and rotate their authorities; Vitalik explains why EIP-8141 matters for scaling, not just account abstraction; Base opens its Vibenet preview network to test 200ms blocks and native AA; and Robinhood Chain keeps running while its data briefly stops reaching Ethereum. ERC-8403 Stand
DEV Community · 1h ago